1. Field of the Invention
The invention relates to a method for managing radio resources in a radio system and a radio system.
2. Background
The amount of users and the density of mobile equipment are further increasing together with the rising need for transferring large amounts of information over wireless communications systems. This causes increasingly tightening demands and requirements for the capacity of mobile communication systems.
It has been proposed that instead of just one type of radio access technology, future wireless communication networks should be using several types of radio access technologies, such as WCDMA (Wideband Code Division Multiple Access), GSM/EDGE (Global System for Mobile Communication/Enhanced Data Rates for Global Evolution) or the like. With the use of different technologies, the network as a whole can take advantage of coverage and capacity characteristics of the different technologies. This causes new demands for the management of network resources. A critical demand will be the managing quality of service (QoS) in a network without wasting resources.
In the near future, wireless communication networks and wireless user equipment will more and more support Internet protocol (IP) based technologies. In most of the fixed networks, the amount of IP traffic has already exceeded that of circuit switched traffic. It also has been proposed that when IP traffic begins to dominate radio access networks (RAN) network layers can be removed and costs reduced by native IP handling.
Managing radio resources is a critical task for future communication networks, especially for communication networks using packet switched transport networks with a routing function, e.g. IP packet based transport networks. IP packet based transport exploits one of the very basic features of IP packets: when user data is packetized by means of IP, an IP packet comprises the information about the packet's destination together with its origin, which makes the packet easily routable. This makes the IP packet very tolerant for network failures or constant topology changes, problems that in other types of networks are usually solved by external logic, such as protection and management systems. The length of the IP packet is variable, which makes it easy to adapt to various types of signalling and user data.
When data is put into IP packet form, the destination address causes specific routing decisions in routers, through which the packet travels on route to its destination. In circuit switched networks, the content is unaware of its destination, and the network signals the connection. Network capacity is reserved by the signalled connection as long as the connection lasts. In packet networks, user data is sent over a specific link and several consequent packets can share the same connection.
Routing was originally designed for file transfer, so it is naturally optimized for that purpose. The above described routing performs well also for real time traffic, such as voice and videotelephony, as long as there is enough network capacity available and the quality of service (QoS) is taken care of.
Problems arise when the communication network is congested, e.g. part of the communication network becomes overloaded. Congestion can occur if, for example, the routers or other network elements receive data faster than the data can be forwarded from the router. If the traffic is allowed to flow freely to the IP transport of the mobile network, like on the Internet, especially the thin transport part close to the base stations may become congested.
In prior art it has been suggested that, when destination routes are congested, data packets are either drop or put on hold, depending for example on the router's capabilities. Packets being queued at buffers in the communication system can be dropped to make room for arriving packets. New packets can be prevented from entering the congested part of the communication system until room for new data is made. However, these techniques cause problems, such as dropped data or delay, and variation of delay that degrade the quality of service and that therefore are unwanted especially in real-time telecommunication services.
From the coverage point of view, it may be necessary to build more radio capacity than there is transport capacity on those links close to the base stations. The problem is that this may even increase the possibility of ending up in a congested situation where e.g. handover from one cell to another would be sensible from the radio point of view but not from the transport point of view. Congestion may be further increased by link failures, which lower the transport capacity.
The data congestion problem is also introduced in a previous application (PCT/IB02/00919) of the applicant. In the application, the problem is solved by data routing: a congested portion of a communication system is identified, and data throughput in the congested portion is reduced by routing data via a selected base station. The application describes an IP transport Resource Manager (ITRM) which manages the transport control, e.g. the quality of service (QoS) in the transport network, by monitoring the IP data flow through the transmission network, and which receives measurements from various elements of the network, indicating, among other things, traffic congestion. An entity called a common resource management server (CRMS) is also described in the application, the CRMS being responsible for the management of radio resource control. The CRMS receives periodical load and QoS measurements from each radio cell of the radio network and acts as an advisor to an entity responsible for the radio network control functions of the radio access network in question. The previous application introduces a way to pass information between the ITMR and the CMRS using messages from the ITMR to the CMRS.